Datasets:
| license: mit | |
| task_categories: | |
| - video-text-to-text | |
| This repository contains the data presented in [MR. Video: "MapReduce" is the Principle for Long Video Understanding](https://huggingface.co/papers/2504.16082). |